Determining Occupancy: Residential Edition

My last post on determining occupancy for LEED projects did not discuss how to determine residential occupancy, an issue I've encountered recently.

Searching through the Credit Interpretation Rulings, I found a rather vague answer coughed up by the supreme court that is the CIR council (committee? emperor?) dated 2/14/2007:

Per the LEED NC v2.1 SSc4.2 ruling dated 12/9/2005, occupancy should be calculated using the number of bedrooms in the case of residential projects. Project teams should provide a narrative with calculations demonstrating how the total number of residential FTE occupants was calculated including indication of number of units, size of units (e.g. studio, one bedroom, two bedroom, etc.), and assumptions as to number of occupants per each size of unit.

So we base it on bedrooms... great! It would sure be nice to have actual guidance on how to transfer that into a hard number, but why make it that simple? Now we get into the sticky issue of what number to put to what bedroom count. Being lazy, I did not take the time to research average occupancy across the nation or anything like that. Instead I proposed the following:

Residential Occupancy Per Bedroom

  • Studio/1 Bedroom - 1.5 occupants
  • 2 Bedrooms - 2.5 occupants
  • 3 Bedrooms - 3.5 occupants
  • 4 Bedrooms - 4.5 occupants
  • etc.

Being the smart guy or gal that you are, you've probably picked up on "the formula". I think it's a safe bet to assume that a one bedroom could equally be expected to be occupied by 1 or 2 people, so I split the difference and assume 1.5. For additional bedrooms, it's likely to be occupied by either kids or roommates, but doubtful it will be occupied by additional couples.

I'd like to be clear that this assumption has not gone off for review yet, and I'm not going to spend a few hundred dollars to answer the question beforehand... I think this argument is reasonable, and expect it to be accepted, but don't want you to stake a large quantity of money on the fact that I'm right. If you have used different assumptions that worked or tried this one and failed, PLEASE be sure to comment on your experience below. Thanks!


Pat said...

Joel. Keep up the good work on your website. I'm starting a LEED project soon and I know I'll be coming to your website often. Thanks for all of the help.

I hope I'm driving some good traffic to your site!

joelmckellar said...

Actually, you're my biggest referral... Thanks!

Anonymous said...

On some of our multifamily projects, we've used one person per bedroom and any room that can be used as a bedroom such as a den. It appears that as long as your formula is reasonable for the particular project, the reviewers allow you to use whatever formula you invent.

Dave Low said...

Does anyone else have any feedback on this?

I have used 1.25 per bedroom, have had it accepted and would like to use it again - but I have also seen a formula similar to Joel's - 2 people for a 1 bedroom, 3 people for a 2 bedroom, 4 people for a 3 bedroom, etc.

Joel - thanks for this resource!

Andy O said...

Has anyone been able to get back to the 2005 ruling? The CIR index seems to cut off before 2005, before v.2.2. Email has been predictably ineffective.

Jafar said...

I am working on a high-rise residential project and I'm using the occupancy densities specified in CaGBC's MURB Guide.

The CaGBC Application Guide for Multi-Unit Residential Buildings (MURB) lists the followings occupancy densities:

1 bedroom/bachelor = 1 occupant
2 bedrooms = 2 occupants
3 bedrooms = 3 occupants

I believe this basis should be acceptable for projects registered under USGBC as well.

Hope that helps!

joelmckellar said...


Thanks for the info! As that's a lower count than I proposed that will probably be helpful for some folks. I'm going to have to track down that guide!


Jimmy Park said...

I have a question.
I am assuming 3 residents per a 2-bed rooms unit. When determining a number of workstations, would you count 2 controls per 3 workstations (3 residents), or 2 controls per 2 workstations (2 bedrooms).
Do you have evidence or experience on this?